PBC is an autoimmune liver disease characterized by intrahepatic bile duct obstruction and antimitochondrial autoantibodies (AMAs). Although considerable insightshave been gained on both the autoantibody response and the CD4+ T cell response, a detailed understanding of the mechanisms of destruction of biliary epithelial cells (BECs) remains elusive. Ons of the highlights of our findings has been the intracellular localization of material selectively in PBC at the apical surface of BECs, which reacts with autoantibodies specific for the major mitochondrial autoantigens, PDC-E2, BCOADC-E2, and OGDC-E2. Such localization prompts us to propose three specific issues. First, because PBC is characterized by extensive CD8+ infiltrates and our pilot data shows a marked increase in PDC-E2 specific MHC class I restricted CD8+ T cells, we will determine the nature of autoantigen specific epitopes recognized by MHC class I restricted CD8+ T cells and identify the immunodominant epitopes. We will also determine the precursor frequency of CD8+ CTLs and, using the immunodominant epitope bearing appropriate MHC class I tetramers, determine whether such frequencies change with disease stage. We also hope to identify differential T cell responses to altered peptide ligands for potential disease specific immunotherapeutic reagents. Second, the immune response in the liver includes a dominant contribution of IgA as well as NK-T cells. Therefore, we will focus on the role of autoantigen-specific IgA in the destruction of BECs using our human recombinant mAb IgA AMAs. Moreover, we will define a role for CD Id using our CDId tetramers in the immune response to the distinct lipoylated mitochondrial autoantigens. Third, the early appearance and abundance of immunoreactivity at the apical surface of BECs, suggests that this accumulation may be a central event in PBC and secondary to abnormal proteolytic processing. Also, given the unique profiles of peptides generated by the reaction of PDC-E2 with caspases 3 and 8 and granzyme B, we will determine if such proteases .generate peptides that react differentially with our panel of apical and non-apical staining mAbs. We submit that there are multiple pathways of destruction of bile ducts and this response is orchestrated by an interplay of CD4+ T helper cells, the induction of cytotoxic T cells and B lymphocytes, including specifically reactive IgA in situ. We submit that this rigorous dissection of PBC will define pathways that will direct us to potential: mmunotherapeutic agents.